Reliable means of detecting changes which occur during the pre-seizure state could serve as a method of seizure prediction, a benchmark in epilepsy research (NIH Curing Epilepsy Conferences, 2000 and 2007). Our preliminary data indicate pre-seizure constriction in brain extracellular space (ECS) accompanied by reduction in near-infrared (NIR) optical scattering prior to detection of seizure by electroencephalography (EEG). The objective in this application is to determine the optical characteristics of the pre-seizure state in clinically relevant animal models of epilepsy. Three specific aims will be pursued: (1) To test the hypothesis that optical signals of the pre-seizure state can be used to predict epileptiform activity in vitro. Our preliminary data indicate that optical coherence tomography (OCT)-derived signals precede epileptiform activity in vitro. In this Aim, we will characterize the optical changes that occur prior to epileptiform activity in vitro in the hippocampal slice using simultaneous high-resolution microelectrode array (MEA) and OCT recordings. These experiments will fully define the optical changes occurring during the pre-seizure state and during epileptiform activity in vitro. (2) To test the hypothesis that optical signals of the pre-seizure state can be used to predict acute seizures in vivo. Our preliminary data indicate that OCT-derived reflectance intensity decreases prior to seizures in vivo (Eberle et al., 2012). In this Aim, we will test the ability of optical signal detection via OCT imaging to detect the pre-seizure state in vivo in well-established models of generalized and focal acute cortical seizures. These experiments will validate the existence of pre-seizure optical changes in distinct seizure models and provide proof-of-concept for the prediction of seizure onset in vivo with optical methods. (3) To test the hypothesis that implanted fiberoptic NIR probes can be used to detect the pre-seizure state of epileptic animals. Our preliminary data indicate that fiberoptics stereotactically implanted in mouse hippocampus demonstrate reduction in NIR reflectance prior to acute seizures in vivo. The gold standard for clinical application would be to reliably detect a spontaneous seizure in an epileptic animal. Therefore, in this Aim we will apply our novel fiberoptic NIRS detection system to a well-established animal model of chronic epilepsy (intrahippocampal kainic acid model). Sensitivity, specificity, and time course of optical NIR reflectance changes before and during chronic spontaneous seizures will be determined. These experiments will provide proof-of-principle for the efficacy of implanted fiberoptic monitoring to detect epileptic seizures for the first time. Our approach is innovative in (i) focusing on optical scattering changes rather than absorption changes as in prior studies; (ii) the first combination of MEA and OCT technologies in vitro and in vivo; (iii) use of novel fiberoptic NIR probes to measure optical changes in deep brain structures prior to seizures in vivo for the first time. The proposed research is significant because the results will elucidate optical characteristics of the pre-seizure state and lead to methods to detect focal and generalized seizures with unprecedented spatiotemporal resolution.